Does managerial sentiment affect accrual estimates? Evidence from the banking industry

Hribar, Paul; Melessa, Samuel Joseph; Small, Rashied; Wilde, Jaron H.

doi:10.1016/j.jacceco.2016.10.001

Cited by 59 publications

(41 citation statements)

References 55 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…15 This magnitude is larger than the effect of future changes in nonperforming loans (ΔNPL i,t+1 ), a previously documented provisions determinant indicating that some bank managers are forward looking (Beatty and Liao, 2011). Hribar et al (2017) document that managerial sentiment biases bank managers' accrual estimates. Both the sentiment measure and the PU index are macro-level variables with only time-series variation and both are shown to be correlated with several other macro-level variables, such as GDP growth and the change in the unemployment rate.…”

Section: Main Findingsmentioning

confidence: 96%

“…Under this alternative view, the sensitivity of loan loss provisions with respect to PU manifests from behavioral bias rather than from the future economic conditions that bank managers expect. For example, Hribar et al (2017) find that managerial sentiment is negatively associated with loan loss provisions in the banking industry and that evidence from future loan charge-offs suggests that loan loss provisions result from the behavioral biases of under-(over-) provisioning in periods of high (low) managerial sentiment.…”

Section: H2 the Association Between Loan Loss Provisions And Pu Willmentioning

confidence: 98%

“…Following prior literature (e.g., Liu and Ryan, 2006;Liao, 2011, 2014;Bushman and Williams, 2012;Hribar et al, 2017), we also control for a series of bank-level characteristics in the above baseline model. First, we include ∆NPL i,t-1 and ∆NPL i,t-2 because loan loss estimation is usually based on the historical trend of nonperforming loans, i.e., the accounting rule currently in effect for credit losses is backward looking.…”

Section: Model and Variablesmentioning

confidence: 99%

“…In the last two columns of Table 2, we further control for managerial sentiment. As a first step, we replicate Hribar et al's (2017) results in column (6). Consistent with their main findings, the coefficient on managerial sentiment (BELIEFS t ) is significantly negative.…”

Section: Main Findingsmentioning

confidence: 99%

“…In addition to economic expectations, two other explanations for high loan loss provisions in times of higher PU are i) managers' attempts to opportunistically shift earnings from uncertain to more certain times (Stein and Wang, 2016) and ii) negative managerial sentiment not supported by underlying loan conditions (Hribar et al, 2017). To provide evidence that making more loan loss provisions during higher PU periods indeed reflects the economic expectations about future loan losses, we further examine the association between PU and borrowers' future defaults and how it varies with the current quarter's loan loss provisions.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Policy uncertainty and loan loss provisions in the banking industry

Saffar

Zhang

2020

Rev Account Stud

View full text Add to dashboard Cite

Policy uncertainty is an increasingly important issue facing many economies. In this paper, we examine how banks accrue for loan losses in response to policy uncertainty (PU) and the implications of these accruals in terms of actual loan losses and future liquidity creation. Consistent with banks recognizing more loan losses in anticipation of PU's depressive effects, we document a contemporaneous positive association between PU and loan loss accruals. This positive association is more pronounced for banks with a riskier loan portfolio and that have a history of lower loan loss reserves. We also find that banks making more loan loss provisions in times of higher PU have significantly higher future loan charge-offs and lower future liquidity creation. Overall, our paper highlights that PU affects the loan loss accruals of banks and that these accruals reflect rational expectations about PU's depressive effects on the economy.

show abstract

Section: Main Findingsmentioning

confidence: 96%

Section: H2 the Association Between Loan Loss Provisions And Pu Willmentioning

confidence: 98%

Section: Model and Variablesmentioning

confidence: 99%

Section: Main Findingsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Policy uncertainty and loan loss provisions in the banking industry

Saffar

Zhang

2020

Rev Account Stud

View full text Add to dashboard Cite

show abstract

Managerial sentiment and predicted and opportunistic special items

Askarzadeh

Yung

Najand

2023

J Corp Accounting Finance

View full text Add to dashboard Cite

Prior studies suggest that firm managers misclassify core expenses as special items to influence investors’ perception of firm performance. In this investigation, we find that managerial decisions regarding misclassification of expenses are affected by managerial sentiment. Managerial sentiment has a negative relation with special‐item reporting because optimistic (pessimistic) managers pursue a misclassification that is lower (higher) than what is justified by the information available to the manager. By decomposing special items into predicted (PREDSI) and opportunistic (OPSI) components, we find that the effect of managerial sentiment is pronounced among opportunistic special items. Our results show that PREDSI (OPSI) has a non‐negative (negative) relation with future firm performance, and special items associated with managerial sentiment adversely affect future firm performance. We also find results that suggest that financial constraints serve as a channel that reinforces the negative association between managerial sentiment and special items. Specifically, financial constraints impose a limit on how frequently sentiment‐driven managers can misclassify core expenses because misclassifications obscure firm transparency and harm the credit supply. Our results remain consistent after a battery of robustness tests.

show abstract